[Pharo-project] XOR: don't recieve a block

Diogenes Moreira diogenes.moreira at gmail.com
Thu Nov 18 19:27:06 CET 2010


Do not know if this matters, but I'm seeing a copy of the definition
of the definition smalltalk ANSI 1997 and xor: should receive both
blocks and Booleans

Ill trying send to you.

On Thu, Nov 18, 2010 at 3:16 PM, Marcus Denker <marcus.denker at inria.fr> wrote:
>
> On Nov 18, 2010, at 6:58 PM, Levente Uzonyi wrote:
>
>> On Thu, 18 Nov 2010, Marcus Denker wrote:
>>
>>>
>>> On Nov 18, 2010, at 6:34 PM, Stéphane Ducasse wrote:
>>>
>>>> & |  don't
>>>> Do not cnfuse lazy and the others.
>>>
>>>
>>> true xor: [false]
>>>
>>> --> walkback.
>>>
>>> false xor: [false]
>>>
>>> --> [false]
>>>
>>> There is a call to #value missing (which will keep it working with booleans, too, as they understand
>>> #value, too (returning self)).
>>
>> That's a possible solution, but xor can't be lazy, so the block is just pointless.
>>
> So the question is if we do it for consistency...
> (between or:, and:, xor:. They are so close that it's strange to have different parameters for them.
>
> On the negative side, it does cost a send of #value (quick return self) for
> the case that one calls it on a boolean...
>
>        Marcus
>
> --
> Marcus Denker  -- http://www.marcusdenker.de
> INRIA Lille -- Nord Europe. Team RMoD.
>
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: standard_v1_9.doc.tar.bz2
Type: application/x-bzip2
Size: 221361 bytes
Desc: not available
URL: <http://lists.gforge.inria.fr/pipermail/pharo-project/attachments/20101118/229b479b/attachment.bin>


More information about the Pharo-project mailing list